Modeling In-flight Inert Gas Distribution in a 747 Center-Wing Fuel Tank William Cavage AAR-440 Fire Safety Branch Wm. J. Hughes Technical Center Federal.

Modeling In-flight Inert Gas Distribution in a 747 Center-Wing

Fuel TankWilliam Cavage

AAR-440 Fire Safety BranchWm. J. Hughes Technical CenterFederal Aviation Administration

International Systems Fire Protection Working Group

CAA House London, England UK

June 23-24, 2005

AAR-440 Fire Safety R&D

Modeling of Fuel Tank Inerting____________________________________

Outline• Background

• Previous Work

• Plywood Scale Test Article

• Analytical Model

• Results– Cascading Inerting

– 747 SCA Test Simulations

• Summary



Background• To support the OBIGGS development work, Fire Safety

Research has been developing models of ullage inerting given inert gas deposited and flight cycle– Assists in the development process and allows for system

development, cost analysis, and trade studies

– Models have to be simple to be useful for these applications

• Would like to be able to model the distribution of inert gas in compartmentalized tanks– Help determine the necessary distribution needed to meet future

requirements for reducing oxygen concentration– Want to capitalize on previous modeling work done in support of

ground-based inerting research (see AIAA Paper 2002-3032)



Previous Work – Multi-Bay GBI Inerting Models• FAA developed analytical model that calculates inert gas

distribution, in terms of [O2], in 6-bay tank at sea level

– Based on previous work and tracks oxygen in and out of each bay at each time step, given inert gas deposited, assumes perfect mixing

• FAA built and performed tests in 24% scale 747SP CWT and measured inert gas distribution– Inerted with scaled flow and measured [O2] in each bay

• Boeing Phantom Works created a CFD Model of a Boeing 747SP center-wing fuel tank and solved for a single, GBI flow case– Used CFD FluentTM Solver and assumed no turbulent mixing

• Results gave varying degrees of agreement with aircraft data– Scale tank had excellent agreement, all duplicated bulk average well


Modeling of Fuel Tank Inerting____________________________________Analytical Model Inerting Data Comparison

0

5

10

15

20

25

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6

Overall VTE

Ox

yg

en

Co

nc

en

tra

tio

n (

% v

ol) Bay 1 Bay 1

Bay 2 Bay 2 Bay 3 Bay 3 Bay 4 Bay 4 Bay 5 Bay 5 Bay 6 Bay 6

B-747 SP CWT Inerting, Single Bay DepositComparison with Analytical Model

B-747 Data Analytical Model

VolumeTankFuel

RateFlowVolumeTimeExchangeTankVolumetric


Modeling of Fuel Tank Inerting____________________________________Scale Model Inerting Data Comparison

0

5

10

15

20

25

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6

Overall VTE

Ox

yg

en

Co

nc

en

tra

tio

n (

% v

ol) Bay 1 Bay 1

Bay 2 Bay 2 Bay 3 Bay 3 Bay 4 Bay 4 Bay 5 Bay 5 Bay 6 Bay 6

B-747 SP CWT Inerting, Single Bay DepositComparison with Scale Tank Data

B-747 Data Scale Tank



0

5

10

15

20

25

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6

Overall VTE

Oxy

gen

Co

nce

ntr

atio

n (

% v

ol)

Bay 1 Bay 1 Bay 2 Bay 2 Bay 3 Bay 3 Bay 4 Bay 4 Bay 5 Bay 5 Bay 6 Bay 6

CWT Inerting, Single Bay DepositComparison with CFD Data

B-747 Data CFD Data

CFD Model Inerting Data Comparison



0

5

10

15

20

25

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8

Volumetric Tank Exchange

Av

era

ge

Ox

yg

en

Co

nc

en

tra

tio

n (

% v

ol)

Full-Scale Test Article

24% Scale Tank

Analytical Model

CFD Analysis

Weighted Average CWT Inerting DataSingle Bay Deposit, 95% NEA, 20-Minute Inerting

8% Line

VolumeTankFuel

RateFlowVolumeTimeExchangeTankVolumetric

CFD Model Inerting Data Comparison



Previous Work – In-flight Tank Inerting Modeling• Developed analytical model of ullage oxygen concentration in a

single bay tank based on inert gas added and altitude change – Model uses system performance (NEA flow and purity) in terms of time

and altitude and calculates mass of oxygen added and removed from one bay at each time step, assuming perfect mixing, and given tank volume and starting oxygen concentration

– Calculates ullage gas removed from tank due to increase in altitude (mass of O2 ) and air entering tank due to decrease in altitude (mass of O2 )

• FAA built and performed tests in 50% scale Airbus A320 center-wing fuel tank using altitude chamber– Made from plywood using drawings given to FAA by Airbus

– Mass flow controller and NEA mixer used to inert the tank with scaled flow in altitude chamber and measured ullage with oxygen analyzer


Modeling of Fuel Tank Inerting____________________________________A320 Flight Test Descent Data Compared with Model Results

0

2

4

6

8

10

12

14

65 70 75 80 85 90 95 100 105

Time (mins)

Oxy

gen

Co

nce

ntr

atio

n (

% v

ol)

0

5,000

10,000

15,000

20,000

25,000

30,000

35,000

40,000

45,000

Alt

itu

de

(1

00

0 f

t)

A320 Flight Test [O2] Analytical Model [O2] Scale Model [O2] Altitude

Comparison of Model Data - Average Ullage [O2]Airbus Test v1972 - Dual Flow System



Scale 747 SCA CWT In-flight Testing• FAA used existing 24% scale model in an altitude chamber

to simulate 747 SCA flight test scenario– Used model from previous GBI experiments

– Test personnel controlled the altitude/time flight cycle to simulate particular flight test

– Inerting system simulated by additional test personnel using mass flow controller and NEA mixer to inert the tank with scaled flow in altitude chamber

– 8-channel altitude oxygen analyzer used to track each bay ullage oxygen concentration as well as additional basic instrumentation to monitor experiment

– See Report DOT/FAA/AR-04/41 for details of 747 SCA flight tests



Bay 1Bay 2

NEA Flow In

Bay n

Air

Ullage

Bay iBay n-1 . . .

• Developed analytical model of multi-bay inerting in-flight based on previous model to simulate 747 SCA flight test scenario

Model Method - Multi Bay Analytical Model

– First developed more simple “cascading” inerting model which has very few assumptions and is more easily validated

– Model has one inert gas deposit (bay 1) and one vent (bay n)

Bay 1

Bay 2

NEA Flow In

Bay 3 Bay 4

Bay 5 Bay 6

Air

Ullage

Air

Ullage

– Next, modified model to split flow to several bays and vent flow from several bays using flight test data for ratios



0

5

10

15

20

25

0 5 10 15 20 25 30 35 40

Time (mins)

Oxy

gen

Co

nce

ntr

atio

n (

% v

ol)

Bay 1 Bay 1 Bay 2 Bay 2 Bay 3 Bay 3 Bay 4 Bay 4

Comparison of Cascading Inerting Test with Model Data

Analytical Model Measured Data

Increasing Altitude

StableCruise

Decreasing Altitude

Results – Cascading Inerting• Reconfigured the 747 scale tank to perform cascading inerting

tests to compare results of scale tank with analytical model– Changed deposit and venting configuration and made holes between bays

small to promote mixing

– Results good but had significant discrepancies during descent

– Differences contributed to scale tank lid leaking air in due to worn seal

– Bulk average data matched identically further supporting conclusion



Results – 747 SCA Ground Test• Used data from a single flight test to tweak analytical model and

make small change to scale tank – Both

models agreed well with ground test results

– Scale tank still had some differences

– Bulk average data matched well

0

5

10

15

20

25

0 10 20 30 40 50 60 70

Time (mins)

Oxy

gen

Co

nce

ntr

atio

n (

% v

ol)

Bay 1 Bay 1 Bay 1

Bay 2 Bay 2 Bay 2

Bay 3 Bay 3 Bay 3Bay 4 Bay 4 Bay 4

Bay 5 Bay 5 Bay 5

Bay 6 Bay 6 Bay 6

Comparison of Model Methods with Flight Test Data - Ground Inerting

Flight Test Scale TankModel



Results – 747 SCA Descent Simulation• Duplicated several descents using both analytical model and

scale tank simulations– Used measured OBIGGS performance and flight cycle as input to scale

tank and analytical model to simulate a test descent

– Results of both simple models illustrated good agreement with the data trends with several large discrepancies in values at individual times, peak and resulting values of the critical bays 1, 2, & 3 had deviations of about +/- 1% oxygen (other bays larger)

– Bulk average of the different models agreed well with measured flight test data

– Limitations of analytical model evident when comparing all flight test descent – simplistic nature of the model gives limited capabilities

– Scale tank needs to be looked at more to determine why it has difficulty in predicting the measured results



0

5

10

15

20

25

0 10 20 30 40 50 60 70

Descent Time (mins)

Oxy

gen

Co

nce

ntr

atio

n (

% v

ol)

Bay 1 Bay 1 Bay 1

Bay 2 Bay 2 Bay 2

Bay 3 Bay 3 Bay 3

Comparison of Model Methods with Flight Test Data - Inerting During Descent


747 SCA Flight Test Descent Data Compared with Model Results



0

2

4

6

8

10

12

14

0 10 20 30 40 50 60 70

Descent Time (mins)

Oxy

gen

Co

nce

ntr

atio

n (

% v

ol)

Bay 4 Bay 4 Bay 4

Bay 5 Bay 5 Bay 5

Bay 6 Bay 6 Bay 6






0

2

4

6

8

10

12

14

16

0 10 20 30 40 50 60 70

Descent Time (mins)

Oxy

gen

Co

nce

ntr

atio

n (

% v

ol)

Scale Tank

Analytical Model

Flight Test *

Average Ullage [O2]Comparison of Model Methods with Flight Test Data - Inerting During Descent

* Average Calculated with Fictitious Bay 4 Data



Modeling of Fuel Tank Inerting____________________________________747 SCA Flight Test Descent Data Compared with Model Results

0

2

4

6

8

10

12

14

16

18

0 10 20 30 40 50 60

Descent Time (mins)

Oxy

gen

Co

nce

ntr

atio

n (

% v

ol)

Bay 1 Bay 1 Bay 1

Bay 2 Bay 2 Bay 2

Bay 3 Bay 3 Bay 3





• An analytical model of ullage oxygen concentration distribution has been developed for a 6-bay compartmentalized tank and can duplicate flight test trend data in a fairly accurate manner given experimental data is available to develop bay-to-bay flow ratios– Peak and resulting values agree fair to good depending on case

• A scale model can be used with an altitude facility in a relatively cost effective way to give fair agreement with peak and resulting flight test [O2] values– Some specialized instrumentation and facilities needed

– More work is needed to see the effect on small scale tank fidelity on the ability of the model to duplicate flight test data

Summary

Modeling In-flight Inert Gas Distribution in a 747 Center-Wing Fuel Tank William Cavage AAR-440 Fire Safety Branch Wm. J. Hughes Technical Center Federal.

Documents

modeling of fuel tank

inerting research

tank volume

single bay tank

previous modeling work

safety rd background

distribution of inert

sp centerwing fuel tank